Abstact. Tuberculosis continues to be a disease of significant morbidity and mortality. Complicating its management are its diverse manifestations, ranging from acute disease to latent infection. It is estimated that 1/3 of the world's population has latent TB, resulting in an enormous reservoir from which reactivation to acute disease can occur. More effective drugs are urgently needed to simplify and shorten treatment courses in order to address the challenges of compliance to the requisite lengthy courses required for sterilization and cure. During both active and latent infection, it has been proposed that a population or subpopulation of bacteria enters a reversible non-replicating state, refractory to traditional antibiotics. The term phenotypic antibiotic tolerance (drug tolerance) is used to describe the reduced efficacy of antibiotics against these bacteria in the absence of genotypic resistance. In Mtb, it has been proposed that in vivo drug tolerance could explain the persistence of infection in the face of prolonged therapy. Thus, integrating our understanding of drug tolerance and the basis for bacterial survival in the face of chemotherapy into therapeutic discovery could be key to designing new strategies for targeting latent and persistent infection. Recently, antibiotic-induced reactive oxygen species (ROS) have been recognized as playing an important role in antibiotic efficacy in susceptible cells while mechanisms that protect against ROS play a role in drug tolerance. In fact, such protective mechanisms have been implicated in all current in vitro models of drug tolerance. Thus, inhibiting or overwhelming the mechanisms that allow drug-tolerant bacteria to survive the cellular stress of ROS is an innovative, promising strategy to rapidly sterilize latent or persistent infection In this project, we propose to develop novel assays to identify small molecule candidates which are able to sterilize drug tolerant Mtb bacilli by disrupting their ability to detoxify and thus survive the stress of ROS, or that increase ROS production in bacilli thus contributing to their death. We will use a model of drug tolerance that exploits the ability of thiourea to quench hydroxyl radicals generated by antibiotic exposure, thereby inducing drug tolerance. We will then screen two unique, valuable collections of small molecules: a collection of 676 molecules that we have already identified as having activity against nutrient starved Mtb which are otherwise drug tolerant to currently available TB drugs, and a unique 100,000 diversity oriented synthetic molecule library that has been created at the Broad Institute. Promising candidates will be developed to obtain molecules for testing in chronic and latent mouse models of TB. This work will integrate the novel biological concept of targeting ROS-mediated mechanisms for targeting persistence and latency with novel chemistry in the DOS library collection, and will draw upon drug discovery expertise from leaders in the pharmaceutical industry now at the Broad Institute and expertise in in vivo testing of TB drug candidates at Johns Hopkins.